You're confusing two very different things. "Pay attention to the user's behavior" and "listen to what the user asks for".
The first is always valuable. Seeing what users do is just plain good. You should be doing that. You should absolutely be doing that.
The second, however, is a frequent mistake. Users don't know what they want. They know what they want to do, and they either know they can't do it or they know how they used to be able to do it, but the ideas they come up with to fix that issue tend to range the gamut from "barely acceptable" to "horrible".
Any change you make to an existing UI - *any change whatsoever* - will result in a storm of people calling for blood. No matter how good the idea is, no matter how good the change is, people will scream for it to be changed back. If you want to create a good UI, at some point you just have to ignore this. People yell for reversion, you tell them "no", and a few months down the line you find out if you made the right call or not.
You might think he made the wrong decision here, but "listening to the users" has absolutely nothing to do with real user experience testing.